Lyn and Fyn function as molecular switches that control immunoreceptors to direct homeostasis or inflammation

Abstract

Immunoreceptors can transduce either inhibitory or activatory signals depending on ligand avidity and phosphorylation status, which is modulated by the protein kinases Lyn and Fyn. Here we show that Lyn and Fyn control immune receptor signaling status. SHP-1 tyrosine 536 phosphorylation by Lyn activates the phosphatase promoting inhibitory signaling through the immunoreceptor. By contrast, Fyn-dependent phosphorylation of SHP-1 serine 591 inactivates the phosphatase, enabling activatory immunoreceptor signaling. These SHP-1 signatures are relevant in vivo, as Lyn deficiency exacerbates nephritis and arthritis in mice, whereas Fyn deficiency is protective. Similarly, Fyn-activating signature is detected in patients with lupus nephritis, underlining the importance of this Lyn-Fyn balance. These data show how receptors discriminate negative from positive signals that respectively result in homeostatic or inflammatory conditions.Src-family kinases Fyn and Lyn are signaling components downstream of ITAM-bearing antigen receptors. Here the authors show that by phosphorylating SHP-1 at different residues, Lyn and Fyn can have opposing regulatory effects on ITAM receptors.

Fig. 1

Differential regulation of FcR-ITAM signals by Lyn and Fyn. a After induction of FcγRIIA-ITAMi or ITAM signalling in THP-1-CD14⁺-FcγRIIA⁺ cells transfected with indicated siRNAs, immunoprecipitation (IP) and immunoblots (IB) were performed with indicated Abs. Quantification of the indicated band using ImageJ software relative to total corresponding protein levels in cell lysates (see Supplementary Fig. 1) is indicated at the bottom of each panel, representing one out of at least three experiments. b Modulation of LPS-mediated IL-8 production by Lyn and Fyn during FcγRIIA-ITAMi induction. THP-1-CD14⁺-FcγRIIA⁺ cells transfected with indicated siRNAs were stimulated for indicated time points to induce either ITAMi or ITAM signals followed by stimulation with LPS (10 ng/ml) for 1 h. Then, supernatant was collected for cytokine measurement. c Modulation of IL-8 production by Lyn and Fyn during FcγRIIA-ITAM induction for 18 h. Data are presented as the mean ± s.e.m. ***P < .001; Student’s unpaired t-test

Fig. 2

Differential regulation of BCR-ITAM signals by Lyn and Fyn. a After induction of BCR-ITAMi or ITAM signalling in transfected Ramos B cells, immunoprecipitation (IP) and immunoblots (IB) were performed with indicated Abs. Quantification of the indicated band using ImageJ software relative to total corresponding protein levels in cell lysates is shown at the bottom of each panel, representing one out of at least three experiments. b Modulation of Pam3csk4-mediated IL-8 production by Lyn and Fyn during BCR-ITAMi signalling on transfected Ramos B cells were stimulated for 30 min to induce ITAMi signal followed by stimulation with Pam3csk4 (1 µg/ml) for 6 h. Supernatants were collected for cytokine measurement. c Modulation of IL-8 production by Lyn and Fyn on transfected Ramos B cells after induction of BCR-ITAM signaling for 6 h. Supernatants were collected for cytokine measurement. For all, data are presented as the mean ± s.e.m. n = 3. ***P < .001; Student’s unpaired t-test. NS, not stimulated

Fig. 3

Fyn–PI3K–PKCα axis inactivates SHP-1-mediated ITAMi signaling. a Modulation of LPS-mediated IL-8 production by Lyn or Fyn after induction of FcγRIIA-ITAM signal on transfected THP-1. Cells were stimulated with LPS for 1 h at 37 °C after induction of ITAM as described in Fig. 1a. Data are presented as the mean ± s.e.m. ***P < .001; Student’s unpaired t-test. b After induction of FcγRIIA-mediated ITAMi or ITAM signals on BMDM derived from FcγRIIA transgenic mice or from FcγRIIA^Tg under Lyn-deficient or Fyn-deficient backgrounds. Cell lysate samples were subjected to SDS-PAGE and immunoblots were performed using anti-phospho (p) SHP-1 serine 591 (S591) or tyrosine 536 (Y536) Abs. c Effect of PI3K, PKC, and ERK inhibitors on SHP-1 phosphorylation driven by ITAM signals in BMDMs. Cells were pre-treated with inhibitors as indicated and Western blotting was performed on cell lysates using antibodies anti-phosphorylated (p) kinases and phosphatases as indicated. Total protein contents in cell lysates were shown in Supplementary Fig. 3,b. d–f Involvement of PKCα on serine SHP-1 phosphorylation following FcγRIIA-ITAM signal. BMDMs from FcγRIIA^Tg (d), Lyn-deficient FcγRIIA^Tg (e), or Fyn-deficient FcγRIIA^Tg (f) mice were transfected with indicated siRNAs before induction of FcγRIIA-ITAM as described in Fig. 1a. b–f Are representative of three experiments. NS, not stimulated

Fig. 4

Lyn-SHP-1^Y536 axis protects mice against lethal nephritis. a Survival curves, b proteinuria, and c serum blood urea nitrogen (BUN) after NTN induction. d Rabbit IgG deposit quantification by immunohistochemistry (IHC) (top panels), haematoxylin & eosin (H&E), and fibrosis Masson’s stain (middle panels) of kidney sections from one representative out of nine mice. Immunostaining for CD11b⁺ and F4/80⁺ cells in kidney sections of the indicated mouse lines (bottom panels). e Relative gene expression of indicated cytokines assessed by q-PCR of independent kidney tissue RNA samples. *P < .05, **P < .01, ***P < .001; Mann–Whitney test. Non significant, ns. Data are from at least five mice per group. f Representative photomicrographs of glomeruli stained for phalloidin and p-Syk-Alexa 647. g Representative photomicrographs of glomeruli stained for phalloidin, p-SHP-1^S591-Alexa 647 and pSHP-1^Y536-Alexa 488. Scale bars: 200 μm

Fig. 5

Fyn axis is essential for arthritis development. a Fyn-deficient mice failed to develop autoimmune arthritis following induction of the CAIA model both in hFcγRIIA^Tg and hFcαRI^Tg background. Arthritis score was graded blind as 0 (normal), 3 (mild), 6 (moderate), or 9 (severe). b, c Circulating levels of CII-specific antibodies in individual sera from WT, FcγRIIA^Tg (b) and from FcαRI^Tg (c) mice on the indicated backgrounds at day 10 after injection of antibodies. Mean ± s.e.m. (n = 4) (d) Lyn is essential for FcR-ITAMi-mediated protection of arthritis. Arthritis score evaluated as in a. e, f H&E staining of hind paws from representative indicated mice at day 10. Scale bars: 200 µm. Mean ± s.e.m (n = 8). **P < .01; two-way ANOVA test

Fig. 6

Inactive SHP-1^S591 is linked to arthritis development. a–c Immunostaining for the detection of phospho SHP-1^Y536 (a), phospho SHP-1^S591 (b), and phospho Syk^Y525–526 (c) in hind leg sections of indicated mice subjected or not to the CAIA model. The corresponding quantifications of positive cells for phospho-SHP-1 on Y⁵³⁶ or S⁵⁹¹ and phospho-Syk Y^525/526 are shown on the right of each panel. Scale bars: 200 µm. **P < .01, ***P < .001; non significant ns, Mann–Whitney test. Sections of three mice/group were automatically quantified with the software CaloPix piloted in a blind manner. Data are presented as the mean ± s.e.m (n = 5)

Fig. 7

Fyn-SHP-1^S591 axis as a biomarker of lupus nephritis activity. a Analysis of phosphorylated ITAM effectors in blood leukocyte samples from 4 lupus nephritis (LN) patients (1 and 2 are class IV; 3 and 4 are class V) and four healthy individuals (HI). Left panel, cell lysates were immunoblotted using indicated Abs. Right panel, FcγRIIA immunoprecipitation (IP) with an anti-FcγRIIA monoclonal antibody (IV.3) followed by immunoblotting using indicated antibodies. b Quantification of the indicated band using ImageJ software relative to total levels of the corresponding protein in cell lysates. *P < .05, **P < .01, ***P < .001; Mann–Whitney test. Not detectable (ND). Data are mean ± s.e.m (n = 5). No significant differences in blood phagocyte counts were observed between healthy controls and patients as followed: Monocytes 0.50 ± 0.16 vs. 0.49 ± 0.30; Neutrophils 4.0 ± 0.42 vs. 3.4 ± 2.4, ×10⁶/ml, respectively. c Representative photomicrographs of glomeruli stained from biopsies of lupus nephritis patients for phalloidin (red), p-SHP-1^S591-Alexa 647 (blue), and pSHP-1^Y536-Alexa 488 (green). Scale bars: 200 μm. d Representative photomicrographs of glomeruli stained from biopsies of LN patients for phalloidin (red), p-SHP-1^S591-Alexa 647 (blue), and CD68-Alexa 405 (white). Scale bars: 200 μm. e Predicted model for the Lyn and Fyn switch controlling the balance between inhibitory or activating ITAM signals in immunoreceptors. Left panel: upon divalent targeting of immunoreceptors, Lyn is recruited to the receptor leading to receptor partial phosphorylation on tyrosine (e.g., Y³⁰⁴ in the case of FcγRIIA). Simultaneously, Lyn phosphorylates SHP-1 on Y⁵³⁶, inducing a conformational change leading to SH2 domain recruitment to phospho-ITAM, thereby lifting inhibition of the phosphatase domain by the N-SH2 domain. This enables SHP-1 to inactivate signal effectors recruited by heterologous receptors. Right panel: multivalent crosslinking of immunoreceptors results in full phosphorylation of ITAM tyrosines by Fyn. These phosphotyrosines serve as “docking” sites for Syk, inducing cell activation and inflammatory responses. Fyn simultaneously activates the PI3K–PKCα pathway, leading to SHP-1 phosphorylation on S⁵⁹¹. The N-SH2/phosphatase domains are maintained in a closed conformation, blocking both recruitment and activation of SHP-1. Under chronic stimulation, this may lead to aggravation of inflammatory or autoimmune diseases